Prediction of adolescent suicidal behavior from established clinical factors is limited by gaps in knowledge about the interplay between distal and proximal risk factors. Persistently-elevated suicide ideation during adolescence is a stable factor associated with increased risk of suicide attempts, but proximal mechanisms that explain how persistent ideation increases risk of or converts to attempts remain untested. This knowledge gap presents a barrier to effective suicide prevention and to developing interventions tailored to adolescents? suicide-ideation-based risk subtypes. In the proposed research, we test how suicide ideation subtypes -- persistent versus brief ? interact with proximal individual differences to drive extended persistent ideation and increase risk of future suicide attempts. We target two individual differences posited to underlie persistent ideation: attention biases (AB) towards suicide-related stimuli and increased interpersonal stress sensitivity (ISS). In addition, we will examine the contribution of disrupted sleep given its association with greater severity of suicide ideation and potential role in amplifying AB and ISS. The proposed study leverages measures of AB and ISS at multiple levels of analysis, including behavioral reaction times, self-report, ecological momentary assessment (EMA) of ideation, scalp-recorded event-related potentials of AB and ISS, and actigraphy measurement of sleep disturbances. Two hundred racially- and economically-diverse adolescents recruited from three New York City hospitals will be classified via interview as persistent versus brief suicide ideators. Aim 1 will be to test whether suicide ideation subtypes differ in biobehavioral profiles of AB, ISS, and sleep disruption. Aim 2 will be to test whether ideation subtype predicts distinct day-to-day patterns of suicide ideation over a one-month follow-up period using EMA, including degree of or change in day-to-day variability in ideation. Aim 3 will be to test the independent and interactive contributions of suicide ideation subtype, biobehavioral individual differences, and day-to-day patterns of ideation in predicting long- term ideation severity and suicide attempts over 3-, 6-, and 12-month follow up. Using an innovative combination of biobehavioral measures, this proposal aims to elucidate both stable and temporally-sensitive patterns of ideation and of proposed proximal factors underlying risk of suicide attempts. Confirmation of hypotheses would represent a crucial step towards identifying individual differences in biobehavioral profiles that distinguish and can be combined with classification of suicide ideation subtypes to predict persistence of suicide ideation and risk of a suicide attempt among adolescents. Ultimately, findings have the potential to speed progress towards tailored interventions to prevent the transition from suicidal thought to suicidal behavior in this vulnerable population.